diff --git a/ggml/src/ggml-metal/ggml-metal.m b/ggml/src/ggml-metal/ggml-metal.m
index 0aec92a26..b7434da3e 100644
--- a/ggml/src/ggml-metal/ggml-metal.m
+++ b/ggml/src/ggml-metal/ggml-metal.m
@@ -2028,7 +2028,7 @@ static void ggml_metal_encode_node(
                        src0t == GGML_TYPE_Q4_K ||
                        src0t == GGML_TYPE_Q5_K ||
                        src0t == GGML_TYPE_Q6_K ||
-                       false) && (ne11 >= 4 && ne11 <= 12)
+                       false) && (ne11 >= 4 && ne11 <= 8)
                      )
                     )
                    ) {
diff --git a/tests/test-backend-ops.cpp b/tests/test-backend-ops.cpp
index 7bd1cb5e0..c786da4c3 100644
--- a/tests/test-backend-ops.cpp
+++ b/tests/test-backend-ops.cpp
@@ -3572,7 +3572,7 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
     test_cases.emplace_back(new test_rwkv_wkv6(GGML_TYPE_F32, 32, 64, 32, 4));
     test_cases.emplace_back(new test_rwkv_wkv6(GGML_TYPE_F32, 32, 64, 128, 4));
 
-    for (int i = 1; i < 64; ++i) {
+    for (int i = 1; i < 9; ++i) {
         test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F16,    GGML_TYPE_F32, 16,  i, 256, { 1,  1}, {1, 1}));
         test_cases.emplace_back(new test_mul_mat(GGML_TYPE_Q4_0,   GGML_TYPE_F32, 16,  i, 256, { 1,  1}, {1, 1}));
         test_cases.emplace_back(new test_mul_mat(GGML_TYPE_Q4_1,   GGML_TYPE_F32, 16,  i, 256, { 1,  1}, {1, 1}));